US9436722B1 - Parallel checksumming of data chunks of a shared data object using a log-structured file system - Google Patents

Parallel checksumming of data chunks of a shared data object using a log-structured file system Download PDF

Info

Publication number
US9436722B1
US9436722B1 US13/799,264 US201313799264A US9436722B1 US 9436722 B1 US9436722 B1 US 9436722B1 US 201313799264 A US201313799264 A US 201313799264A US 9436722 B1 US9436722 B1 US 9436722B1
Authority
US
United States
Prior art keywords
log
file system
data
data chunk
structured file
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active, expires
Application number
US13/799,264
Inventor
John M. Bent
Sorin Faibish
Gary Grider
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
EMC Corp
Triad National Security LLC
Original Assignee
EMC Corp
Los Alamos National Security LLC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by EMC Corp, Los Alamos National Security LLC filed Critical EMC Corp
Priority to US13/799,264 priority Critical patent/US9436722B1/en
Priority to US13/799,228 priority patent/US9477682B1/en
Assigned to EMC CORPORATION reassignment EMC CORPORATION ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: FAIBISH, SORIN, BENT, JOHN M.
Assigned to LOS ALAMOS NATIONAL SECURITY, LLC reassignment LOS ALAMOS NATIONAL SECURITY, LLC ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: GRIDER, GARY A.
Assigned to U.S. DEPARTMENT OF ENERGY reassignment U.S. DEPARTMENT OF ENERGY CONFIRMATORY LICENSE (SEE DOCUMENT FOR DETAILS). Assignors: LOS ALAMOS NATIONAL SECURITY
Publication of US9436722B1 publication Critical patent/US9436722B1/en
Application granted granted Critical
Assigned to TRIAD NATIONAL SECURITY, LLC reassignment TRIAD NATIONAL SECURITY, LLC ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: LOS ALAMOS NATIONAL SECURITY, LLC
Assigned to THE BANK OF NEW YORK MELLON TRUST COMPANY, N.A. reassignment THE BANK OF NEW YORK MELLON TRUST COMPANY, N.A. SECURITY AGREEMENT Assignors: CREDANT TECHNOLOGIES, INC., DELL INTERNATIONAL L.L.C., DELL MARKETING L.P., DELL PRODUCTS L.P., DELL USA L.P., EMC CORPORATION, EMC IP Holding Company LLC, FORCE10 NETWORKS, INC., WYSE TECHNOLOGY L.L.C.
Assigned to THE BANK OF NEW YORK MELLON TRUST COMPANY, N.A. reassignment THE BANK OF NEW YORK MELLON TRUST COMPANY, N.A. SECURITY AGREEMENT Assignors: CREDANT TECHNOLOGIES INC., DELL INTERNATIONAL L.L.C., DELL MARKETING L.P., DELL PRODUCTS L.P., DELL USA L.P., EMC CORPORATION, EMC IP Holding Company LLC, FORCE10 NETWORKS, INC., WYSE TECHNOLOGY L.L.C.
Assigned to EMC IP Holding Company LLC, DELL PRODUCTS L.P., DELL INTERNATIONAL L.L.C., DELL MARKETING CORPORATION (SUCCESSOR-IN-INTEREST TO FORCE10 NETWORKS, INC. AND WYSE TECHNOLOGY L.L.C.), EMC CORPORATION, DELL MARKETING L.P. (ON BEHALF OF ITSELF AND AS SUCCESSOR-IN-INTEREST TO CREDANT TECHNOLOGIES, INC.), DELL USA L.P. reassignment EMC IP Holding Company LLC RELEASE OF SECURITY INTEREST IN PATENTS PREVIOUSLY RECORDED AT REEL/FRAME (053546/0001) Assignors: THE BANK OF NEW YORK MELLON TRUST COMPANY, N.A., AS NOTES COLLATERAL AGENT
Active legal-status Critical Current
Adjusted expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/10—File systems; File servers
    • G06F16/18—File system types
    • G06F16/1858—Parallel file systems, i.e. file systems supporting multiple processors
    • G06F17/30371—
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/23—Updating
    • G06F16/2365—Ensuring data consistency and integrity
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
    • G06F3/067—Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/04—Protocols for data compression, e.g. ROHC

Definitions

  • the present invention relates to parallel storage in high performance computing environments.
  • Parallel storage systems are widely used in many computing environments. Parallel storage systems provide high degrees of concurrency in which many distributed processes within a parallel application simultaneously access a shared file namespace.
  • Parallel computing techniques are used in many industries and applications for implementing computationally intensive models or simulations.
  • the Department of Energy uses a large number of distributed compute nodes tightly coupled into a supercomputer to model physics experiments.
  • parallel computing techniques are often used for computing geological models that help predict the location of natural resources.
  • each parallel process generates a portion, referred to as a data chunk, of a shared data object.
  • Checksumming is a common technique to ensure data integrity.
  • a checksum or hash sum is a fixed-size computed from a block of digital data to detect errors that may have been introduced during transmission or storage. The integrity of the data can be checked at any later time by recomputing the checksum and comparing the recomputed checksum with the stored checksum. If the two checksum values match, then the data was likely not altered.
  • Embodiments of the present invention provide improved techniques for generating checksum values and for verifying the integrity of data.
  • a method is provided for a client executing on one or more of a compute node and a burst buffer node in a parallel computing system to store a data chunk generated by the parallel computing system to a shared data object on a storage node in the parallel computing system.
  • the client determines a checksum value for the data chunk; and provides the checksum value with the data chunk to the storage node that stores the shared object.
  • the data chunk can be stored on the storage node with the corresponding checksum value as part of the shared object.
  • the storage node may be part of a Parallel Log-Structured File System (PLFS), and the client may comprise, for example, a Log-Structured File System client executing on a compute node or a burst buffer node.
  • PLFS Parallel Log-Structured File System
  • the checksum value can be evaluated when the data chunk is read from the storage node to verify the integrity of the data that is read.
  • illustrative embodiments of the invention provide techniques for parallel checksumming of data being written to a shared object.
  • FIG. 1 illustrates an exemplary conventional technique for generating checksums of data being stored to a shared object by a plurality of processes in a storage system
  • FIG. 2 illustrates an exemplary distributed technique for generating checksums of data being stored to a shared object by a plurality of processes in a storage system in accordance with aspects of the present invention
  • FIG. 3 illustrates an exemplary alternate distributed technique for generating checksums of data being stored to a shared object by a plurality of processes in a storage system in accordance with an alternate embodiment of the present invention
  • FIG. 4 is a flow chart describing an exemplary LSFS checksum process incorporating aspects of the present invention.
  • the present invention provides improved techniques for cooperative parallel writing of data to a shared object.
  • one aspect of the present invention leverages the parallelism of concurrent writes to a shared object and the high interconnect speed of parallel supercomputer networks to generate the checksum values for the data in parallel as it is written.
  • a further aspect of the invention leverages the parallel supercomputer networks to provide improved techniques for verifying the integrity of the checksummed data.
  • Embodiments of the present invention will be described herein with reference to exemplary computing systems and data storage systems and associated servers, computers, storage units and devices and other processing devices. It is to be appreciated, however, that embodiments of the invention are not restricted to use with the particular illustrative system and device configurations shown. Moreover, the phrases “computing system” and “data storage system” as used herein are intended to be broadly construed, so as to encompass, for example, private or public cloud computing or storage systems, as well as other types of systems comprising distributed virtual infrastructure. However, a given embodiment may more generally comprise any arrangement of one or more processing devices. As used herein, the term “files” shall include complete files and portions of files, such as sub-files or shards.
  • FIG. 1 illustrates an exemplary conventional storage system 100 that employs a conventional technique for generating checksums of data being stored to a shared object 150 by a plurality of processes.
  • the exemplary storage system 100 may be implemented, for example, as a Parallel Log-Structured File System (PLFS) to make placement decisions automatically, as described in U.S. patent application Ser. No. 13/536,331, filed Jun. 28, 2012, entitled “Storing Files in a Parallel Computing System Using List-Based Index to Identify Replica Files,” (now U.S. Pat. No. 9,087,075), incorporated by reference herein, or it can be explicitly controlled by the application and administered by a storage daemon.
  • PLFS Parallel Log-Structured File System
  • the exemplary storage system 100 comprises a plurality of compute nodes 110 - 1 through 110 -N (collectively, compute nodes 110 ) where a distributed application process generates a corresponding portion 120 - 1 through 120 -N of a distributed shared data structure 150 or other information to store.
  • the compute nodes 110 optionally store the portions 120 of the distributed data structure 150 in one or more nodes of the exemplary storage system 100 , such as an exemplary flash based storage node 140 .
  • the exemplary hierarchical storage tiering system 100 optionally comprises one or more hard disk drives (not shown).
  • the compute nodes 110 send their distributed data chunks 120 into a single file 150 .
  • the single file 150 is striped into file system defined blocks, and then each block is checksummed.
  • existing checksum approaches apply checksums on the shared data structure 150 only after it has been sent to the storage node 140 of the storage system 100 .
  • the checksums 160 are applied to offset ranges on the data in sizes that are pre-defined by the file system 100 .
  • the offset size of the checksums 160 does not typically align with the size of the data portions 120 (i.e., the file system defined blocks will typically not match the original memory layout).
  • FIG. 2 illustrates an exemplary storage system 200 that generates checksums of data chunks 220 being stored to a shared object 250 by a plurality of processes in accordance with aspects of the present invention.
  • the exemplary storage system 200 may be implemented, for example, as a Parallel Log-Structured File System.
  • the exemplary storage system 200 comprises a plurality of compute nodes 210 - 1 through 210 -N (collectively, compute nodes 110 ) where a distributed application process generates a corresponding data chunk portion 220 - 1 through 220 -N (collectively, data chunks 220 ) of a distributed shared data object 250 to store.
  • the distributed application executing on given compute node 210 in the parallel computing system 200 writes and reads the data chunks 220 that are part of the shared data object 250 using a log-structured file system (LSFS) client 205 - 1 through 205 -N executing on the given compute node 210 .
  • LSFS log-structured file system
  • the compute nodes 210 store the data chunk portions 220 of the distributed data structure 250 in one or more storage nodes of the exemplary storage system 200 , such as an exemplary LSFS server 240 .
  • the LSFS server 240 may be implemented, for example, as a flash based storage node.
  • the exemplary hierarchical storage tiering system 200 optionally comprises one or more hard disk drives (not shown).
  • each LSFS client 205 applies a checksum function to each data chunk 220 to generate a corresponding checksum value 260 - 1 through 260 -N.
  • Each data chunk 220 is then stored by the corresponding LSFS client 205 with the corresponding computed checksum 260 on the LSFS server 240 .
  • the LSFS client 205 performs a data integrity check on the read operation, where the data chunk 220 and the corresponding checksum 260 are read from the LSFS server 240 and are provided to the corresponding LSFS client 205 on the compute node 210 for the data integrity check before being sent to the application.
  • the data integrity check comprises recomputing the checksum 260 recompute and comparing the recomputed checksum 260 recompute with the stored checksum 260 stored . If the two checksum values 260 recompute and 260 stored match, then the data integrity is verified.
  • FIG. 3 illustrates an exemplary storage system 300 that generates checksums of data chunks 220 being stored to a shared object 250 by a plurality of processes in accordance with an alternate embodiment of the present invention.
  • the exemplary storage system 300 may be implemented, for example, as a Parallel Log-Structured File System.
  • the exemplary storage system 300 comprises a plurality of compute nodes 210 - 1 through 210 -N (collectively, compute nodes 110 ) where a distributed application process generates a corresponding data chunk portion 220 - 1 through 220 -N (collectively, data chunks 220 ) of a distributed shared data object 250 to store, in a similar manner to FIG. 2 .
  • the distributed application executing on given compute node 210 in the parallel computing system 200 writes and reads the data chunks 220 that are part of the shared data object 250 using a log-structured file system (LSFS) client 205 - 1 through 205 -N executing on the given compute node 210 , in a similar manner to FIG. 2 .
  • the compute nodes 210 store the data chunk portions 220 of the distributed data structure 250 in one or more storage nodes of the exemplary storage system 200 , such as an exemplary LSFS server 240 .
  • the LSFS server 240 may be implemented, for example, as a flash based storage node.
  • the exemplary hierarchical storage tiering system 200 optionally comprises one or more hard disk drives (not shown).
  • the exemplary storage system 300 also comprises one or more flash-based burst buffer nodes 310 - 1 through 310 - k that process the data chunks 220 that are written by the LSFS clients 205 to the LSFS server 240 , and are read by the LSFS clients 205 from the LSFS server 240 .
  • the exemplary flash-based burst buffer nodes 310 comprise LSFS clients 305 in a similar manner to the LSFS clients 205 of FIG. 2 .
  • each burst buffer node 310 applies a checksum function to each data chunk 220 to generate a corresponding checksum value 360 - 1 through 360 -N.
  • Each data chunk 220 is then stored with the corresponding computed checksum 360 on the LSFS server 240 , in a similar manner to FIG. 2 .
  • the burst buffer node 310 performs a data integrity check on the read operation, where the data chunk 220 and the corresponding checksum 360 are read from the LSFS server 240 and are provided to the the burst buffer node 310 for the data integrity check before being sent to the application executing on the compute node 210 .
  • the data integrity check comprises recomputing the checksum 360 recompute and comparing the recomputed checksum 360 recompute with the stored checksum 360 stored . If the two checksum values 360 recompute and 360 stored match, then the data integrity is verified.
  • FIGS. 2 and 3 can be combined such that checksumming is performed by the LSFS clients 205 executing on the compute nodes 210 and additional more computationally intensive checksumming and/or erasure coding is performed by the burst buffer nodes 310 .
  • a distributed signature lookup service can be established across the network of burst buffer nodes 310 to reduce the latency to verify the checksums.
  • FIG. 4 is a flow chart describing an exemplary LSFS checksum process 400 incorporating aspects of the present invention.
  • the exemplary LSFS checksum process 400 is implemented by the LSFS clients 205 executing on the compute nodes 210 in the embodiment of FIG. 2 and by the flash-based burst buffer nodes 310 in the embodiment of FIG. 3 .
  • the exemplary LSFS checksum process 400 initially performs a test during step 410 to determine if the current operation is a read operation or a write operation. If it is determined during step 410 that the current operation is a write operation, then the exemplary LSFS checksum process 400 obtains the data chunk from the application during step 420 . The exemplary LSFS checksum process 400 then computes the checksum for the data chunk during step 430 on the compute nodes 210 or the burst buffer nodes 310 . Finally, the data chunk is stored on the LSFS server 240 with the corresponding checksum as part of the shared object 250 during step 440 .
  • a test is performed during step 470 to determine if the checksums match. If it is determined during step 470 that the checksums match, then the verified data chunk is provided to the application on the compute node 210 during step 480 . If, however, it is determined during step 470 that the checksums do not match, then the exemplary LSFS checksum process 400 indicates a failure to the application or corrects the error and provides corrected data during step 490 . For example, when the block being read does not match exactly a block that was checksummed but is comprised of pieces from several blocks, the burst buffer layer 310 can check the checksums from the multiple blocks and recompute a new checksum for the block being read and then send just the block and the checksum to the compute server 210 .
  • the number of compute servers 210 is at least an order of magnitude greater than the number of storage servers 240 in HPC systems, thus it is much faster to perform the checksum computations on the compute servers 210 .
  • the checksumming is performed on the data chunks 220 as they are being written by the LSFS client 205 as opposed to when they have been placed into the file 250 by the server 240 .
  • the chunks 220 in a log-structured file system retain their original data organization whereas in existing approaches, the data in the chunks will almost always be reorganized into file system defined blocks. This can introduce additional latency as the file system will either wait for the blocks to be filled or do the checksumming multiple times each time the block is partially filled.
  • aspects of the present invention leverage the parallelism of concurrent writes to a shared object and the high interconnect speed of parallel supercomputer networks to improve the generation of checksums during a write operation and to use the checksummed data to improve the data integrity on read operations.
  • aspects of the present invention thus recognize that the log-structured file system eliminates the need for artificial file system boundaries because all block sizes perform equally well in a log-structured file system.
  • PLFS files can be shared across many locations, data processing required to implement these functions can be performed more efficiently when there are multiple nodes cooperating on the data processing operations. Therefore, when this is run on a parallel system with a parallel language, such as MPI, PLFS can provide MPI versions of these functions which will allow it to exploit parallelism for more efficient data processing.
  • the storage server node 240 can also optionally check the checksum as a mechanism to detect data corruption during the network transmission on a write operation from the compute nodes 210 to the storage server node 240 . In the event the checksum computed by the storage server node 240 does not match the checksum received from the compute node 210 with the data chunk 220 , the data can be re-transmitted to obtain the uncorrupted data.
  • Such components can communicate with other elements over any type of network, such as a wide area network (WAN), a local area network (LAN), a satellite network, a telephone or cable network, or various portions or combinations of these and other types of networks.
  • WAN wide area network
  • LAN local area network
  • satellite network a satellite network
  • telephone or cable network a telephone or cable network

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer Security & Cryptography (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Debugging And Monitoring (AREA)

Abstract

Checksum values are generated and used to verify the data integrity. A client executing in a parallel computing system stores a data chunk to a shared data object on a storage node in the parallel computing system. The client determines a checksum value for the data chunk; and provides the checksum value with the data chunk to the storage node that stores the shared object. The data chunk can be stored on the storage node with the corresponding checksum value as part of the shared object. The storage node may be part of a Parallel Log-Structured File System (PLFS), and the client may comprise, for example, a Log-Structured File System client on a compute node or burst buffer. The checksum value can be evaluated when the data chunk is read from the storage node to verify the integrity of the data that is read.

Description

STATEMENT OF GOVERNMENT RIGHTS
This invention was made under a Cooperative Research and Development Agreement between EMC Corporation and Los Alamos National Security, LLC. The United States government has rights in this invention pursuant to Contract No. DE-AC52-06NA25396 between the United States Department of Energy and Los Alamos National Security, LLC for the operation of Los Alamos National Laboratory.
CROSS-REFERENCE TO RELATED APPLICATION
This application is related to U.S. patent application Ser. No. 13/799,228, filed Mar. 13, 2013, entitled “Parallel Compression of Data Chunks of a Shared Object Using a Log-Structured File System,” incorporated by reference herein
FIELD
The present invention relates to parallel storage in high performance computing environments.
BACKGROUND
Parallel storage systems are widely used in many computing environments. Parallel storage systems provide high degrees of concurrency in which many distributed processes within a parallel application simultaneously access a shared file namespace.
Parallel computing techniques are used in many industries and applications for implementing computationally intensive models or simulations. For example, the Department of Energy uses a large number of distributed compute nodes tightly coupled into a supercomputer to model physics experiments. In the oil and gas industry, parallel computing techniques are often used for computing geological models that help predict the location of natural resources. Generally, each parallel process generates a portion, referred to as a data chunk, of a shared data object.
Checksumming is a common technique to ensure data integrity. A checksum or hash sum is a fixed-size computed from a block of digital data to detect errors that may have been introduced during transmission or storage. The integrity of the data can be checked at any later time by recomputing the checksum and comparing the recomputed checksum with the stored checksum. If the two checksum values match, then the data was likely not altered.
Existing approaches apply checksums on the shared data object after it has been sent to the storage system. The checksums are applied to offset ranges on the shared data object in sizes that are pre-defined by the file system.
In parallel computing systems, such as High Performance Computing (HPC) applications, the inherently complex and large datasets increase the potential for data corruption and therefore the need for data integrity. A need therefore exists for parallel techniques for generating the checksum values and for verifying the integrity of the data.
SUMMARY
Embodiments of the present invention provide improved techniques for generating checksum values and for verifying the integrity of data. In one embodiment, a method is provided for a client executing on one or more of a compute node and a burst buffer node in a parallel computing system to store a data chunk generated by the parallel computing system to a shared data object on a storage node in the parallel computing system. The client determines a checksum value for the data chunk; and provides the checksum value with the data chunk to the storage node that stores the shared object. The data chunk can be stored on the storage node with the corresponding checksum value as part of the shared object.
The storage node may be part of a Parallel Log-Structured File System (PLFS), and the client may comprise, for example, a Log-Structured File System client executing on a compute node or a burst buffer node. The checksum value can be evaluated when the data chunk is read from the storage node to verify the integrity of the data that is read.
Advantageously, illustrative embodiments of the invention provide techniques for parallel checksumming of data being written to a shared object. These and other features and advantages of the present invention will become more readily apparent from the accompanying drawings and the following detailed description.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 illustrates an exemplary conventional technique for generating checksums of data being stored to a shared object by a plurality of processes in a storage system;
FIG. 2 illustrates an exemplary distributed technique for generating checksums of data being stored to a shared object by a plurality of processes in a storage system in accordance with aspects of the present invention;
FIG. 3 illustrates an exemplary alternate distributed technique for generating checksums of data being stored to a shared object by a plurality of processes in a storage system in accordance with an alternate embodiment of the present invention; and
FIG. 4 is a flow chart describing an exemplary LSFS checksum process incorporating aspects of the present invention.
 DETAILED DESCRIPTION
The present invention provides improved techniques for cooperative parallel writing of data to a shared object. Generally, one aspect of the present invention leverages the parallelism of concurrent writes to a shared object and the high interconnect speed of parallel supercomputer networks to generate the checksum values for the data in parallel as it is written. A further aspect of the invention leverages the parallel supercomputer networks to provide improved techniques for verifying the integrity of the checksummed data.
Embodiments of the present invention will be described herein with reference to exemplary computing systems and data storage systems and associated servers, computers, storage units and devices and other processing devices. It is to be appreciated, however, that embodiments of the invention are not restricted to use with the particular illustrative system and device configurations shown. Moreover, the phrases “computing system” and “data storage system” as used herein are intended to be broadly construed, so as to encompass, for example, private or public cloud computing or storage systems, as well as other types of systems comprising distributed virtual infrastructure. However, a given embodiment may more generally comprise any arrangement of one or more processing devices. As used herein, the term “files” shall include complete files and portions of files, such as sub-files or shards.
FIG. 1 illustrates an exemplary conventional storage system 100 that employs a conventional technique for generating checksums of data being stored to a shared object 150 by a plurality of processes. The exemplary storage system 100 may be implemented, for example, as a Parallel Log-Structured File System (PLFS) to make placement decisions automatically, as described in U.S. patent application Ser. No. 13/536,331, filed Jun. 28, 2012, entitled “Storing Files in a Parallel Computing System Using List-Based Index to Identify Replica Files,” (now U.S. Pat. No. 9,087,075), incorporated by reference herein, or it can be explicitly controlled by the application and administered by a storage daemon.
As shown in FIG. 1, the exemplary storage system 100 comprises a plurality of compute nodes 110-1 through 110-N (collectively, compute nodes 110) where a distributed application process generates a corresponding portion 120-1 through 120-N of a distributed shared data structure 150 or other information to store. The compute nodes 110 optionally store the portions 120 of the distributed data structure 150 in one or more nodes of the exemplary storage system 100, such as an exemplary flash based storage node 140. In addition, the exemplary hierarchical storage tiering system 100 optionally comprises one or more hard disk drives (not shown).
As shown in FIG. 1, the compute nodes 110 send their distributed data chunks 120 into a single file 150. The single file 150 is striped into file system defined blocks, and then each block is checksummed. As indicated above, existing checksum approaches apply checksums on the shared data structure 150 only after it has been sent to the storage node 140 of the storage system 100. Thus, as shown in FIG. 1, the checksums 160 are applied to offset ranges on the data in sizes that are pre-defined by the file system 100. The offset size of the checksums 160 does not typically align with the size of the data portions 120 (i.e., the file system defined blocks will typically not match the original memory layout).
FIG. 2 illustrates an exemplary storage system 200 that generates checksums of data chunks 220 being stored to a shared object 250 by a plurality of processes in accordance with aspects of the present invention. The exemplary storage system 200 may be implemented, for example, as a Parallel Log-Structured File System.
As shown in FIG. 2, the exemplary storage system 200 comprises a plurality of compute nodes 210-1 through 210-N (collectively, compute nodes 110) where a distributed application process generates a corresponding data chunk portion 220-1 through 220-N (collectively, data chunks 220) of a distributed shared data object 250 to store. The distributed application executing on given compute node 210 in the parallel computing system 200 writes and reads the data chunks 220 that are part of the shared data object 250 using a log-structured file system (LSFS) client 205-1 through 205-N executing on the given compute node 210. The compute nodes 210 store the data chunk portions 220 of the distributed data structure 250 in one or more storage nodes of the exemplary storage system 200, such as an exemplary LSFS server 240. The LSFS server 240 may be implemented, for example, as a flash based storage node. In addition, the exemplary hierarchical storage tiering system 200 optionally comprises one or more hard disk drives (not shown).
In accordance with one aspect of the present invention, on a write operation, each LSFS client 205 applies a checksum function to each data chunk 220 to generate a corresponding checksum value 260-1 through 260-N. Each data chunk 220 is then stored by the corresponding LSFS client 205 with the corresponding computed checksum 260 on the LSFS server 240.
In accordance with another aspect of the present invention, on a read operation, the LSFS client 205 performs a data integrity check on the read operation, where the data chunk 220 and the corresponding checksum 260 are read from the LSFS server 240 and are provided to the corresponding LSFS client 205 on the compute node 210 for the data integrity check before being sent to the application. The data integrity check comprises recomputing the checksum 260 recompute and comparing the recomputed checksum 260 recompute with the stored checksum 260 stored. If the two checksum values 260 recompute and 260 stored match, then the data integrity is verified.
FIG. 3 illustrates an exemplary storage system 300 that generates checksums of data chunks 220 being stored to a shared object 250 by a plurality of processes in accordance with an alternate embodiment of the present invention. The exemplary storage system 300 may be implemented, for example, as a Parallel Log-Structured File System. As shown in FIG. 3, the exemplary storage system 300 comprises a plurality of compute nodes 210-1 through 210-N (collectively, compute nodes 110) where a distributed application process generates a corresponding data chunk portion 220-1 through 220-N (collectively, data chunks 220) of a distributed shared data object 250 to store, in a similar manner to FIG. 2. The distributed application executing on given compute node 210 in the parallel computing system 200 writes and reads the data chunks 220 that are part of the shared data object 250 using a log-structured file system (LSFS) client 205-1 through 205-N executing on the given compute node 210, in a similar manner to FIG. 2. The compute nodes 210 store the data chunk portions 220 of the distributed data structure 250 in one or more storage nodes of the exemplary storage system 200, such as an exemplary LSFS server 240. The LSFS server 240 may be implemented, for example, as a flash based storage node. In addition, the exemplary hierarchical storage tiering system 200 optionally comprises one or more hard disk drives (not shown).
The exemplary storage system 300 also comprises one or more flash-based burst buffer nodes 310-1 through 310-k that process the data chunks 220 that are written by the LSFS clients 205 to the LSFS server 240, and are read by the LSFS clients 205 from the LSFS server 240. The exemplary flash-based burst buffer nodes 310 comprise LSFS clients 305 in a similar manner to the LSFS clients 205 of FIG. 2.
In accordance with one aspect of the present invention, on a write operation, each burst buffer node 310 applies a checksum function to each data chunk 220 to generate a corresponding checksum value 360-1 through 360-N. Each data chunk 220 is then stored with the corresponding computed checksum 360 on the LSFS server 240, in a similar manner to FIG. 2.
In accordance with another aspect of the present invention, on a read operation, the burst buffer node 310 performs a data integrity check on the read operation, where the data chunk 220 and the corresponding checksum 360 are read from the LSFS server 240 and are provided to the the burst buffer node 310 for the data integrity check before being sent to the application executing on the compute node 210. The data integrity check comprises recomputing the checksum 360 recompute and comparing the recomputed checksum 360 recompute with the stored checksum 360 stored. If the two checksum values 360 recompute and 360 stored match, then the data integrity is verified.
On a burst buffer node 310, due to the bursty nature of the workloads, there is additional time to run computationally intensive checksumming as well as addition parity schemes, such as erasure codes, such that corrupted data can be recovered in addition to just being detected.
It is noted that the embodiments of FIGS. 2 and 3 can be combined such that checksumming is performed by the LSFS clients 205 executing on the compute nodes 210 and additional more computationally intensive checksumming and/or erasure coding is performed by the burst buffer nodes 310. A distributed signature lookup service can be established across the network of burst buffer nodes 310 to reduce the latency to verify the checksums.
While these checksums may reduce performance due to latency, this is outweighed by the improved reliability obtained from the additional data integrity. Additionally, on the burst buffer nodes 310, this additional latency will not be incurred by the application since the latency will be added not between the application on the compute nodes 210 and the burst buffer nodes 310 but between the asynchronous transfer from the burst buffer nodes 310 to the lower storage servers 240.
FIG. 4 is a flow chart describing an exemplary LSFS checksum process 400 incorporating aspects of the present invention. The exemplary LSFS checksum process 400 is implemented by the LSFS clients 205 executing on the compute nodes 210 in the embodiment of FIG. 2 and by the flash-based burst buffer nodes 310 in the embodiment of FIG. 3.
As shown in FIG. 4, the exemplary LSFS checksum process 400 initially performs a test during step 410 to determine if the current operation is a read operation or a write operation. If it is determined during step 410 that the current operation is a write operation, then the exemplary LSFS checksum process 400 obtains the data chunk from the application during step 420. The exemplary LSFS checksum process 400 then computes the checksum for the data chunk during step 430 on the compute nodes 210 or the burst buffer nodes 310. Finally, the data chunk is stored on the LSFS server 240 with the corresponding checksum as part of the shared object 250 during step 440.
If, however, it is determined during step 410 that the current operation is a read operation, then the exemplary LSFS checksum process 400 obtains the data chunk and corresponding stored checksum from the LSFS server 240 during step 450. The checksum is recomputed for the data chunk during step 455 and the checksums are compared during step 460 (checksumrecompute=checksumstored?).
A test is performed during step 470 to determine if the checksums match. If it is determined during step 470 that the checksums match, then the verified data chunk is provided to the application on the compute node 210 during step 480. If, however, it is determined during step 470 that the checksums do not match, then the exemplary LSFS checksum process 400 indicates a failure to the application or corrects the error and provides corrected data during step 490. For example, when the block being read does not match exactly a block that was checksummed but is comprised of pieces from several blocks, the burst buffer layer 310 can check the checksums from the multiple blocks and recompute a new checksum for the block being read and then send just the block and the checksum to the compute server 210.
Among other benefits, the number of compute servers 210 is at least an order of magnitude greater than the number of storage servers 240 in HPC systems, thus it is much faster to perform the checksum computations on the compute servers 210. In addition, the checksumming is performed on the data chunks 220 as they are being written by the LSFS client 205 as opposed to when they have been placed into the file 250 by the server 240. The chunks 220 in a log-structured file system retain their original data organization whereas in existing approaches, the data in the chunks will almost always be reorganized into file system defined blocks. This can introduce additional latency as the file system will either wait for the blocks to be filled or do the checksumming multiple times each time the block is partially filled.
In this manner, aspects of the present invention leverage the parallelism of concurrent writes to a shared object and the high interconnect speed of parallel supercomputer networks to improve the generation of checksums during a write operation and to use the checksummed data to improve the data integrity on read operations. Aspects of the present invention thus recognize that the log-structured file system eliminates the need for artificial file system boundaries because all block sizes perform equally well in a log-structured file system.
Because PLFS files can be shared across many locations, data processing required to implement these functions can be performed more efficiently when there are multiple nodes cooperating on the data processing operations. Therefore, when this is run on a parallel system with a parallel language, such as MPI, PLFS can provide MPI versions of these functions which will allow it to exploit parallelism for more efficient data processing.
Consider a partial read. For example, assume that a write operation wrote bytes {0-100} and a checksum, C1, was stored with these bytes as computed at write time. If the reader reads bytes {25-75}, then the checksum, C1, does not match those bytes 25-75. So the storage server node 240 can send the entire byte range or recompute a new checksum, C1′, for bytes 25-75. The former approach has the disadvantage of sending unnecessary data across the network. The latter approach has the disadvantage of doing the checksum computations on the storage server node 240 instead of the much more scalable compute nodes 210 or burst buffer nodes 310.
In a further variation, the storage server node 240 can also optionally check the checksum as a mechanism to detect data corruption during the network transmission on a write operation from the compute nodes 210 to the storage server node 240. In the event the checksum computed by the storage server node 240 does not match the checksum received from the compute node 210 with the data chunk 220, the data can be re-transmitted to obtain the uncorrupted data.
CONCLUSION
Numerous other arrangements of servers, computers, storage devices or other components are possible. Such components can communicate with other elements over any type of network, such as a wide area network (WAN), a local area network (LAN), a satellite network, a telephone or cable network, or various portions or combinations of these and other types of networks.
It should again be emphasized that the above-described embodiments of the invention are presented for purposes of illustration only. Many variations may be made in the particular arrangements shown. For example, although described in the context of particular system and device configurations, the techniques are applicable to a wide variety of other types of information processing systems, data storage systems, processing devices and distributed virtual infrastructure arrangements. In addition, any simplifying assumptions made above in the course of describing the illustrative embodiments should also be viewed as exemplary rather than as requirements or limitations of the invention. Numerous other alternative embodiments within the scope of the appended claims will be readily apparent to those skilled in the art.

Claims (20)

What is claimed is:
1. A method performed by a Log-Structured File System client executing on one or more of a compute node and a burst buffer node in a parallel computing system for storing a data chunk generated by said parallel computing system to a shared data object on a storage node in said parallel computing system, comprising:
processing said data chunk generated by said parallel computing system using said Log-Structured File System client, wherein said Log-Structured File System client executes on one or more of said compute node and said burst buffer node in said parallel computing system;
determining, by said Log-Structured File System client, a checksum value for said data chunk as said data chunk is written by said Log-Structured File System client to said storage node, wherein a plurality of other Log-Structured File System clients determine a plurality of checksums for a corresponding plurality of other data chunks of said shared data object in parallel; and
providing, by said Log-Structured File System client, said checksum value with said data chunk to said storage node for storage as part of said shared object, wherein said storage node stores said data chunk with said corresponding checksum value as part of said shared object with said plurality of checksums for said corresponding plurality of other data chunks of said shared data object, and wherein said shared object is shared by a plurality of distributed processes in said parallel computing system.
2. The method of claim 1, wherein said storage node comprises a Log-Structured File server.
3. The method of claim 1, wherein said storage node is part of a log structured virtual parallel file system.
4. The method of claim 1, wherein said storage node is part of a Parallel Log-Structured File System (PLFS).
5. The method of claim 1, further comprising the step of evaluating said checksum value when said data chunk is read from said storage node.
6. The method of claim 5, further comprising the step of recomputing said checksum value when said data chunk is read from said storage node and comparing said recomputed checksum value to said stored checksum value.
7. The method of claim 6, further comprising the step of providing a verified data chunk if said corresponding checksum values match.
8. The method of claim 6, further comprising the step of correcting said data chunk if said corresponding checksum values do not match.
9. A tangible machine-readable recordable storage medium for storing a data chunk generated by a parallel computing system to a shared data object on a storage node in said parallel computing system, wherein one or more software programs when executed by one or more processing devices implement the steps of the method of claim 1.
10. A compute node apparatus in a parallel computing system for executing a Log-Structured File System client that stores a data chunk generated by said parallel computing system to a shared data object on a storage node in said parallel computing system, said compute node apparatus comprising:
a memory; and
at least one hardware device operatively coupled to the memory and configured to:
process said data chunk generated by said parallel computing system using said Log-Structured File System client, wherein said Log-Structured File System client executes on one or more of said compute node and said burst buffer node in said parallel computing system;
determine, by said Log-Structured File System client, a checksum value for said data chunk as said data chunk is written by said Log-Structured File System client to said storage node, wherein a plurality of other Log-Structured File System clients determine a plurality of checksums for a corresponding plurality of other data chunks of said shared data object in parallel; and
provide, by said Log-Structured File System client, said checksum value with said data chunk to said storage node for storage as part of said shared object, wherein said storage node stores said data chunk with said corresponding checksum value as part of said shared object with said plurality of checksums for said corresponding plurality of other data chunks of said shared data object, and wherein said shared object is shared by a plurality of distributed processes in said parallel computing system.
11. The apparatus of claim 10, wherein said storage node comprises a Log-Structured File server.
12. The apparatus of claim 10, wherein said apparatus comprises one or more of a compute node and a burst buffer node.
13. The apparatus of claim 10, wherein said at least one hardware device is further configured to evaluate said checksum value when said data chunk is read from said storage node.
14. The apparatus of claim 13, wherein said at least one hardware device is further configured to recompute said checksum value when said data chunk is read from said storage node and comparing said recomputed checksum value to said stored checksum value.
15. The apparatus of claim 14, wherein said at least one hardware device is further configured to provide a verified data chunk if said corresponding checksum values match.
16. The apparatus of claim 14, wherein said at least one hardware device is further configured to correct said data chunk if said corresponding checksum values do not match.
17. A method performed by a Log-Structured File System server node in a parallel computing system for storing a data chunk as a shared object, comprising:
receiving, by said Log-Structured File System server, said data chunk with a corresponding checksum value from a Log-Structured File System client on a compute node in said parallel computing system, wherein said data chunk is generated by said parallel computing system using said Log-Structured File System client, and wherein said corresponding checksum value is generated by said Log-Structured File System client as said data chunk is written by said Log-Structured File System Client, wherein a plurality of other Log-Structured File System Clients determine a plurality of checksums for a corresponding plurality of other data chunks of said shared data object in parallel; and
storing, by said Log-Structured File System server, said received data chunk and said corresponding checksum value to a shared data object on said storage node, wherein said storage node stores said data chunk with said corresponding checksum value as part of said shared object with said plurality of checksums for said corresponding plurality of other data chunks of said shared data object, and wherein said shared object is shared by a plurality of distributed processes in said parallel computing system.
18. A tangible machine-readable recordable storage medium for storing a data chunk as a shared object in a parallel computing system, wherein one or more software programs when executed by one or more processing devices implement the steps of the method of claim 17.
19. A Log-Structured File System server apparatus in a parallel computing system for storing a data chunk as a shared object, said storage node apparatus comprising:
a memory; and
at least one hardware device operatively coupled to the memory and configured to:
receive, by said Log-Structured File System server, said data chunk with a corresponding checksum value from a Log-Structured File System client on a compute node in said parallel computing system, wherein said data chunk is generated by said parallel computing system using Log-Structured File System client, and wherein said corresponding checksum value is generated by said Log-Structured File System client as said data chunk is written by said Log-Structured File System Client, wherein a plurality of other log-structured file system clients determine a plurality of checksums for a corresponding plurality of other data chunks of said shared data object in parallel; and
store, by said Log-Structured File System server, said received data chunk and said corresponding checksum value to a shared data object on said storage node, wherein said storage node stores said data chunk with said corresponding checksum value as part of said shared object with said plurality of checksums for said corresponding plurality of other data chunks of said shared data object, and wherein said shared object is shared by a plurality of distributed processes in said parallel computing system.
20. The method of claim 17, further comprising the step of providing said data chunk and said corresponding checksum value to a Log-Structured File System client when said data chunk is read.
US13/799,264 2013-03-13 2013-03-13 Parallel checksumming of data chunks of a shared data object using a log-structured file system Active 2034-04-16 US9436722B1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US13/799,264 US9436722B1 (en) 2013-03-13 2013-03-13 Parallel checksumming of data chunks of a shared data object using a log-structured file system
US13/799,228 US9477682B1 (en) 2013-03-13 2013-03-13 Parallel compression of data chunks of a shared data object using a log-structured file system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US13/799,264 US9436722B1 (en) 2013-03-13 2013-03-13 Parallel checksumming of data chunks of a shared data object using a log-structured file system
US13/799,228 US9477682B1 (en) 2013-03-13 2013-03-13 Parallel compression of data chunks of a shared data object using a log-structured file system

Publications (1)

Publication Number Publication Date
US9436722B1 true US9436722B1 (en) 2016-09-06

Family

ID=56878232

Family Applications (2)

Application Number Title Priority Date Filing Date
US13/799,228 Active 2034-04-26 US9477682B1 (en) 2013-03-13 2013-03-13 Parallel compression of data chunks of a shared data object using a log-structured file system
US13/799,264 Active 2034-04-16 US9436722B1 (en) 2013-03-13 2013-03-13 Parallel checksumming of data chunks of a shared data object using a log-structured file system

Family Applications Before (1)

Application Number Title Priority Date Filing Date
US13/799,228 Active 2034-04-26 US9477682B1 (en) 2013-03-13 2013-03-13 Parallel compression of data chunks of a shared data object using a log-structured file system

Country Status (1)

Country Link
US (2) US9477682B1 (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170024158A1 (en) * 2015-07-21 2017-01-26 Arm Limited Method of and apparatus for generating a signature representative of the content of an array of data
US9749418B2 (en) * 2015-08-06 2017-08-29 Koc University Efficient dynamic proofs of retrievability
US10194156B2 (en) 2014-07-15 2019-01-29 Arm Limited Method of and apparatus for generating an output frame
CN109614037A (en) * 2018-11-16 2019-04-12 新华三技术有限公司成都分公司 Data routing inspection method, apparatus and distributed memory system
WO2022043409A1 (en) * 2020-08-28 2022-03-03 Siemens Aktiengesellschaft Computer-implemented method for storing a dataset and computer network
US11399015B2 (en) 2019-06-11 2022-07-26 Bank Of America Corporation Data security tool
US20240111657A1 (en) * 2022-04-27 2024-04-04 Microsoft Technology Licensing, Llc Automatic correctness validation of database management systems
US12353279B2 (en) 2023-08-02 2025-07-08 International Business Machines Corporation Offloading a validity check for a data block from a server to be performed by a client node

Families Citing this family (49)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10411959B1 (en) 2014-12-30 2019-09-10 EMC IP Holding Company LLC Data analytics for the internet of things
US10761743B1 (en) 2017-07-17 2020-09-01 EMC IP Holding Company LLC Establishing data reliability groups within a geographically distributed data storage environment
US10817388B1 (en) 2017-07-21 2020-10-27 EMC IP Holding Company LLC Recovery of tree data in a geographically distributed environment
US10880040B1 (en) 2017-10-23 2020-12-29 EMC IP Holding Company LLC Scale-out distributed erasure coding
US10382554B1 (en) 2018-01-04 2019-08-13 Emc Corporation Handling deletes with distributed erasure coding
US10817374B2 (en) 2018-04-12 2020-10-27 EMC IP Holding Company LLC Meta chunks
US11301165B2 (en) 2018-04-26 2022-04-12 International Business Machines Corporation Accelerating shared file checkpoint with local burst buffers
US10579297B2 (en) 2018-04-27 2020-03-03 EMC IP Holding Company LLC Scaling-in for geographically diverse storage
US11023130B2 (en) 2018-06-15 2021-06-01 EMC IP Holding Company LLC Deleting data in a geographically diverse storage construct
US10936196B2 (en) 2018-06-15 2021-03-02 EMC IP Holding Company LLC Data convolution for geographically diverse storage
US11436203B2 (en) 2018-11-02 2022-09-06 EMC IP Holding Company LLC Scaling out geographically diverse storage
US10901635B2 (en) 2018-12-04 2021-01-26 EMC IP Holding Company LLC Mapped redundant array of independent nodes for data storage with high performance using logical columns of the nodes with different widths and different positioning patterns
US11119683B2 (en) * 2018-12-20 2021-09-14 EMC IP Holding Company LLC Logical compaction of a degraded chunk in a geographically diverse data storage system
US10931777B2 (en) 2018-12-20 2021-02-23 EMC IP Holding Company LLC Network efficient geographically diverse data storage system employing degraded chunks
US10892782B2 (en) 2018-12-21 2021-01-12 EMC IP Holding Company LLC Flexible system and method for combining erasure-coded protection sets
US11520742B2 (en) 2018-12-24 2022-12-06 Cloudbrink, Inc. Data mesh parallel file system caching
US11023331B2 (en) 2019-01-04 2021-06-01 EMC IP Holding Company LLC Fast recovery of data in a geographically distributed storage environment
US10768840B2 (en) 2019-01-04 2020-09-08 EMC IP Holding Company LLC Updating protection sets in a geographically distributed storage environment
US10942827B2 (en) 2019-01-22 2021-03-09 EMC IP Holding Company LLC Replication of data in a geographically distributed storage environment
US10936239B2 (en) 2019-01-29 2021-03-02 EMC IP Holding Company LLC Cluster contraction of a mapped redundant array of independent nodes
US10942825B2 (en) 2019-01-29 2021-03-09 EMC IP Holding Company LLC Mitigating real node failure in a mapped redundant array of independent nodes
US10846003B2 (en) 2019-01-29 2020-11-24 EMC IP Holding Company LLC Doubly mapped redundant array of independent nodes for data storage
US10866766B2 (en) 2019-01-29 2020-12-15 EMC IP Holding Company LLC Affinity sensitive data convolution for data storage systems
US11029865B2 (en) 2019-04-03 2021-06-08 EMC IP Holding Company LLC Affinity sensitive storage of data corresponding to a mapped redundant array of independent nodes
US10944826B2 (en) 2019-04-03 2021-03-09 EMC IP Holding Company LLC Selective instantiation of a storage service for a mapped redundant array of independent nodes
US11113146B2 (en) 2019-04-30 2021-09-07 EMC IP Holding Company LLC Chunk segment recovery via hierarchical erasure coding in a geographically diverse data storage system
US11121727B2 (en) 2019-04-30 2021-09-14 EMC IP Holding Company LLC Adaptive data storing for data storage systems employing erasure coding
US11119686B2 (en) 2019-04-30 2021-09-14 EMC IP Holding Company LLC Preservation of data during scaling of a geographically diverse data storage system
US11748004B2 (en) 2019-05-03 2023-09-05 EMC IP Holding Company LLC Data replication using active and passive data storage modes
US11209996B2 (en) 2019-07-15 2021-12-28 EMC IP Holding Company LLC Mapped cluster stretching for increasing workload in a data storage system
US11023145B2 (en) 2019-07-30 2021-06-01 EMC IP Holding Company LLC Hybrid mapped clusters for data storage
US11449399B2 (en) 2019-07-30 2022-09-20 EMC IP Holding Company LLC Mitigating real node failure of a doubly mapped redundant array of independent nodes
US11228322B2 (en) 2019-09-13 2022-01-18 EMC IP Holding Company LLC Rebalancing in a geographically diverse storage system employing erasure coding
US11449248B2 (en) 2019-09-26 2022-09-20 EMC IP Holding Company LLC Mapped redundant array of independent data storage regions
US11288139B2 (en) 2019-10-31 2022-03-29 EMC IP Holding Company LLC Two-step recovery employing erasure coding in a geographically diverse data storage system
US11119690B2 (en) 2019-10-31 2021-09-14 EMC IP Holding Company LLC Consolidation of protection sets in a geographically diverse data storage environment
US11435910B2 (en) 2019-10-31 2022-09-06 EMC IP Holding Company LLC Heterogeneous mapped redundant array of independent nodes for data storage
US11144207B2 (en) * 2019-11-07 2021-10-12 International Business Machines Corporation Accelerating memory compression of a physically scattered buffer
US11435957B2 (en) 2019-11-27 2022-09-06 EMC IP Holding Company LLC Selective instantiation of a storage service for a doubly mapped redundant array of independent nodes
US11144220B2 (en) 2019-12-24 2021-10-12 EMC IP Holding Company LLC Affinity sensitive storage of data corresponding to a doubly mapped redundant array of independent nodes
US11231860B2 (en) 2020-01-17 2022-01-25 EMC IP Holding Company LLC Doubly mapped redundant array of independent nodes for data storage with high performance
US11507308B2 (en) 2020-03-30 2022-11-22 EMC IP Holding Company LLC Disk access event control for mapped nodes supported by a real cluster storage system
US11288229B2 (en) 2020-05-29 2022-03-29 EMC IP Holding Company LLC Verifiable intra-cluster migration for a chunk storage system
US11693983B2 (en) 2020-10-28 2023-07-04 EMC IP Holding Company LLC Data protection via commutative erasure coding in a geographically diverse data storage system
US11875036B2 (en) 2021-01-13 2024-01-16 Samsung Electronics Co., Ltd. Computing system including host and storage system and having increased write performance
US11847141B2 (en) 2021-01-19 2023-12-19 EMC IP Holding Company LLC Mapped redundant array of independent nodes employing mapped reliability groups for data storage
US11625174B2 (en) 2021-01-20 2023-04-11 EMC IP Holding Company LLC Parity allocation for a virtual redundant array of independent disks
US11449234B1 (en) 2021-05-28 2022-09-20 EMC IP Holding Company LLC Efficient data access operations via a mapping layer instance for a doubly mapped redundant array of independent nodes
US11354191B1 (en) 2021-05-28 2022-06-07 EMC IP Holding Company LLC Erasure coding in a large geographically diverse data storage system

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5953352A (en) * 1997-06-23 1999-09-14 Micron Electronics, Inc. Method of checking data integrity for a raid 1 system
US20030226139A1 (en) * 2002-05-28 2003-12-04 Sheng Lee System update protocol
US6952797B1 (en) * 2000-10-25 2005-10-04 Andy Kahn Block-appended checksums
US20060123250A1 (en) * 1999-07-16 2006-06-08 Intertrust Technologies Corporation Trusted storage systems and methods
US20080282105A1 (en) * 2007-05-10 2008-11-13 Deenadhayalan Veera W Data integrity validation in storage systems
US20090183056A1 (en) * 2008-01-16 2009-07-16 Bluearc Uk Limited Validating Objects in a Data Storage system
US8862561B1 (en) * 2012-08-30 2014-10-14 Google Inc. Detecting read/write conflicts

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8819288B2 (en) * 2007-09-14 2014-08-26 Microsoft Corporation Optimized data stream compression using data-dependent chunking
WO2009062029A1 (en) * 2007-11-09 2009-05-14 Carnegie Mellon University Efficient high performance system for writing data from applications to a safe file system
US9104617B2 (en) * 2008-11-13 2015-08-11 International Business Machines Corporation Using accelerators in a hybrid architecture for system checkpointing
US8849877B2 (en) * 2010-08-31 2014-09-30 Datadirect Networks, Inc. Object file system
US20120089781A1 (en) * 2010-10-11 2012-04-12 Sandeep Ranade Mechanism for retrieving compressed data from a storage cloud
US9619430B2 (en) * 2012-02-24 2017-04-11 Hewlett Packard Enterprise Development Lp Active non-volatile memory post-processing
US9792182B2 (en) * 2013-01-31 2017-10-17 Hewlett Packard Enterprise Development Lp Checkpoint generation

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5953352A (en) * 1997-06-23 1999-09-14 Micron Electronics, Inc. Method of checking data integrity for a raid 1 system
US20060123250A1 (en) * 1999-07-16 2006-06-08 Intertrust Technologies Corporation Trusted storage systems and methods
US6952797B1 (en) * 2000-10-25 2005-10-04 Andy Kahn Block-appended checksums
US20030226139A1 (en) * 2002-05-28 2003-12-04 Sheng Lee System update protocol
US20080282105A1 (en) * 2007-05-10 2008-11-13 Deenadhayalan Veera W Data integrity validation in storage systems
US20090183056A1 (en) * 2008-01-16 2009-07-16 Bluearc Uk Limited Validating Objects in a Data Storage system
US8862561B1 (en) * 2012-08-30 2014-10-14 Google Inc. Detecting read/write conflicts

Non-Patent Citations (7)

* Cited by examiner, † Cited by third party
Title
Dai et al., "ELF: An Efficient Log Structured Flash File System for Micro Sensor Nodes", ACM SensSys, Baltimore, MD (2004). *
Dai et al., "Elf: An Efficient Log Structured Flash File System for Micro Sensor Nodes", ACM SenSys, Baltimore, MD (2004).
Hartman et al., "The Zebra Striped Network File System", ACM Transactions on Computer Systems (1994). *
John Bent, Garth Gibson, Gary Grider, Ben McClelland, Paul Nowoczynski, James Nunez, Milo Polte, Meghan Wingate, "SC '09 Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis", Nov. 14, 2009, ACM New York, NY, USA © 2009 Article No. 21. *
John H. Hartman, "The Zebra Striped Network File System", A dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy in Computer Science in the Graduate Division of the University of Califomia at Berkeley (1994). *
John H. Hartman, "The Zebra Striped Network File System", A dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy in Computer Science in the Graduate Division of the University of California at Berkeley (1994).
Los Alamos National Laboratory, "PLFS: Parallel Log Structured File System", Jan. 14, 2009. *

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10194156B2 (en) 2014-07-15 2019-01-29 Arm Limited Method of and apparatus for generating an output frame
US20170024158A1 (en) * 2015-07-21 2017-01-26 Arm Limited Method of and apparatus for generating a signature representative of the content of an array of data
US10832639B2 (en) * 2015-07-21 2020-11-10 Arm Limited Method of and apparatus for generating a signature representative of the content of an array of data
US9749418B2 (en) * 2015-08-06 2017-08-29 Koc University Efficient dynamic proofs of retrievability
CN109614037A (en) * 2018-11-16 2019-04-12 新华三技术有限公司成都分公司 Data routing inspection method, apparatus and distributed memory system
US11399015B2 (en) 2019-06-11 2022-07-26 Bank Of America Corporation Data security tool
WO2022043409A1 (en) * 2020-08-28 2022-03-03 Siemens Aktiengesellschaft Computer-implemented method for storing a dataset and computer network
US20240111657A1 (en) * 2022-04-27 2024-04-04 Microsoft Technology Licensing, Llc Automatic correctness validation of database management systems
US12353279B2 (en) 2023-08-02 2025-07-08 International Business Machines Corporation Offloading a validity check for a data block from a server to be performed by a client node

Also Published As

Publication number Publication date
US9477682B1 (en) 2016-10-25

Similar Documents

Publication Publication Date Title
US9436722B1 (en) Parallel checksumming of data chunks of a shared data object using a log-structured file system
US9244623B1 (en) Parallel de-duplication of data chunks of a shared data object using a log-structured file system
US10951236B2 (en) Hierarchical data integrity verification of erasure coded data in a distributed computing system
US10394634B2 (en) Drive-based storage scrubbing
US9298386B2 (en) System and method for improved placement of blocks in a deduplication-erasure code environment
US8386835B2 (en) System and method for end-to-end data integrity in a network file system
CN112749039B (en) Method, apparatus and program product for data writing and data recovery
US8370297B2 (en) Approach for optimizing restores of deduplicated data
US9323765B1 (en) Scalable cloud file system with efficient integrity checks
US8468423B2 (en) Data verification using checksum sidefile
US10666435B2 (en) Multi-tenant encryption on distributed storage having deduplication and compression capability
US10581602B2 (en) End-to-end checksum in a multi-tenant encryption storage system
US20110113313A1 (en) Buffer transfer check on variable length data
WO2019001521A1 (en) Data storage method, storage device, client and system
CN109074295B (en) Data recovery with authenticity
JP2019533397A (en) Device and associated method for encoding and decoding data for erasure codes
CN104408154A (en) Repeated data deletion method and device
US9767139B1 (en) End-to-end data integrity in parallel storage systems
CN104375905A (en) Incremental backing up method and system based on data block
US9256503B2 (en) Data verification
US10402262B1 (en) Fencing for zipheader corruption for inline compression feature system and method
US10243583B2 (en) CPU error remediation during erasure code encoding
US9098446B1 (en) Recovery of corrupted erasure-coded data files
US11243890B2 (en) Compressed data verification
JP2013050836A (en) Storage system, method for checking data integrity, and program

Legal Events

Date Code Title Description
AS Assignment

Owner name: EMC CORPORATION, MASSACHUSETTS

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:BENT, JOHN M.;FAIBISH, SORIN;SIGNING DATES FROM 20130422 TO 20130521;REEL/FRAME:030641/0466

AS Assignment

Owner name: LOS ALAMOS NATIONAL SECURITY, LLC, NEW MEXICO

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:GRIDER, GARY A.;REEL/FRAME:031671/0220

Effective date: 20131123

AS Assignment

Owner name: U.S. DEPARTMENT OF ENERGY, DISTRICT OF COLUMBIA

Free format text: CONFIRMATORY LICENSE;ASSIGNOR:LOS ALAMOS NATIONAL SECURITY;REEL/FRAME:032363/0115

Effective date: 20131212

STCF Information on status: patent grant

Free format text: PATENTED CASE

AS Assignment

Owner name: TRIAD NATIONAL SECURITY, LLC, NEW MEXICO

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:LOS ALAMOS NATIONAL SECURITY, LLC;REEL/FRAME:047485/0323

Effective date: 20181101

AS Assignment

Owner name: THE BANK OF NEW YORK MELLON TRUST COMPANY, N.A., T

Free format text: SECURITY AGREEMENT;ASSIGNORS:CREDANT TECHNOLOGIES, INC.;DELL INTERNATIONAL L.L.C.;DELL MARKETING L.P.;AND OTHERS;REEL/FRAME:049452/0223

Effective date: 20190320

Owner name: THE BANK OF NEW YORK MELLON TRUST COMPANY, N.A., TEXAS

Free format text: SECURITY AGREEMENT;ASSIGNORS:CREDANT TECHNOLOGIES, INC.;DELL INTERNATIONAL L.L.C.;DELL MARKETING L.P.;AND OTHERS;REEL/FRAME:049452/0223

Effective date: 20190320

MAFP Maintenance fee payment

Free format text: PAYMENT OF MAINTENANCE FEE, 4TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1551); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

Year of fee payment: 4

AS Assignment

Owner name: THE BANK OF NEW YORK MELLON TRUST COMPANY, N.A., TEXAS

Free format text: SECURITY AGREEMENT;ASSIGNORS:CREDANT TECHNOLOGIES INC.;DELL INTERNATIONAL L.L.C.;DELL MARKETING L.P.;AND OTHERS;REEL/FRAME:053546/0001

Effective date: 20200409

AS Assignment

Owner name: DELL MARKETING L.P. (ON BEHALF OF ITSELF AND AS SUCCESSOR-IN-INTEREST TO CREDANT TECHNOLOGIES, INC.), TEXAS

Free format text: RELEASE OF SECURITY INTEREST IN PATENTS PREVIOUSLY RECORDED AT REEL/FRAME (053546/0001);ASSIGNOR:THE BANK OF NEW YORK MELLON TRUST COMPANY, N.A., AS NOTES COLLATERAL AGENT;REEL/FRAME:071642/0001

Effective date: 20220329

Owner name: DELL INTERNATIONAL L.L.C., TEXAS

Free format text: RELEASE OF SECURITY INTEREST IN PATENTS PREVIOUSLY RECORDED AT REEL/FRAME (053546/0001);ASSIGNOR:THE BANK OF NEW YORK MELLON TRUST COMPANY, N.A., AS NOTES COLLATERAL AGENT;REEL/FRAME:071642/0001

Effective date: 20220329

Owner name: DELL PRODUCTS L.P., TEXAS

Free format text: RELEASE OF SECURITY INTEREST IN PATENTS PREVIOUSLY RECORDED AT REEL/FRAME (053546/0001);ASSIGNOR:THE BANK OF NEW YORK MELLON TRUST COMPANY, N.A., AS NOTES COLLATERAL AGENT;REEL/FRAME:071642/0001

Effective date: 20220329

Owner name: DELL USA L.P., TEXAS

Free format text: RELEASE OF SECURITY INTEREST IN PATENTS PREVIOUSLY RECORDED AT REEL/FRAME (053546/0001);ASSIGNOR:THE BANK OF NEW YORK MELLON TRUST COMPANY, N.A., AS NOTES COLLATERAL AGENT;REEL/FRAME:071642/0001

Effective date: 20220329

Owner name: EMC CORPORATION, MASSACHUSETTS

Free format text: RELEASE OF SECURITY INTEREST IN PATENTS PREVIOUSLY RECORDED AT REEL/FRAME (053546/0001);ASSIGNOR:THE BANK OF NEW YORK MELLON TRUST COMPANY, N.A., AS NOTES COLLATERAL AGENT;REEL/FRAME:071642/0001

Effective date: 20220329

Owner name: DELL MARKETING CORPORATION (SUCCESSOR-IN-INTEREST TO FORCE10 NETWORKS, INC. AND WYSE TECHNOLOGY L.L.C.), TEXAS

Free format text: RELEASE OF SECURITY INTEREST IN PATENTS PREVIOUSLY RECORDED AT REEL/FRAME (053546/0001);ASSIGNOR:THE BANK OF NEW YORK MELLON TRUST COMPANY, N.A., AS NOTES COLLATERAL AGENT;REEL/FRAME:071642/0001

Effective date: 20220329

Owner name: EMC IP HOLDING COMPANY LLC, TEXAS

Free format text: RELEASE OF SECURITY INTEREST IN PATENTS PREVIOUSLY RECORDED AT REEL/FRAME (053546/0001);ASSIGNOR:THE BANK OF NEW YORK MELLON TRUST COMPANY, N.A., AS NOTES COLLATERAL AGENT;REEL/FRAME:071642/0001

Effective date: 20220329

MAFP Maintenance fee payment

Free format text: PAYMENT OF MAINTENANCE FEE, 8TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1552); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

Year of fee payment: 8